debt into growth: how sovereign debt accelerated the first industrial revolution

31 Pages·2016·0.68 MB·English

DEBT INTO GROWTH: HOW SOVEREIGN DEBT
ACCELERATED THE FIRST INDUSTRIAL REVOLUTION
JAUMEVENTURAANDHANS-JOACHIMVOTH*
Abstract. Why did the country that borrowed the most industrialize first?
Earlier research has viewed the explosion of debt in 18th century Britain as
eitherdetrimental,orasneutralforeconomicgrowth. Inthispaper,weargue
thatthemassiveissuanceofgovernmentbondsallowedthenobilitytoswitch
outoflow-returninvestmentssuchasagriculturalimprovements. Thisinitially
sloweddowngrowth,butthisswitchalsoloweredfactordemandbyoldsectors
andincreasedprofitsinnew,risingonessuchastextilesandiron. Becauseex-
ternalfinancingintheseindustrieswasscarce,thisboostinprofitsaccelerated
growth,andtheIndustrialRevolution. Theabsenceofaneffectivetransferof
financial resources from old to new sectors also helps to explain why the In-
dustrialRevolutionledtomassivesocialchange–becausetherichnobilitydid
notlendtoorinvestintherevolutionizingindustries,itfailedtocapturethe
highreturnstocapitalinthesesectors,leadingtorelativeeconomicdecline.
JEL codes: E22, E25, E62, H56, H60, N13, N23,
Keywords: crowding out, debt crises, Industrial Revolution, Ricardian equivalence, mis-
allocation, financial repression, structural change, productivity
Date:February2016.
*Ventura: CREi,UPF,andBarcelonaGSE.RamonTriasFargas25-27,E-08005Barcelona,jven-
[email protected]. Voth: UniversityofZurich. Schönberggasse1,CH-8001Zurich,[email protected].
WethankKlausAdam,AlbertoAlesina,RolandBenabou,AntonioCiccone,RogerFarmer,Raquel
Fernandez,JordiGalí,OlegItskhoki,ChadJones,PeteKlenow,PeterKoudijs,TimLeunig,Ben-
jaminMoll,JohnvanReenen,KjetlStoresletten,GillesSaintPaul,RomanStuder,NicoVoigtlän-
der,JosephZeira,andFabrizioZilibottiforhelpfulcomments;seminaraudiencesinStanfordGSB,
CREi,theUniversityofOslo,WarwickUniversity,the2014BarcelonaSummerForum,the2015
NBER Summer Institute, and ESSIM-Tarragona 2015 provided helpful feedback. Ventura ac-
knowledgessupportfromtheSpanishMinistryofScienceandInnovation(ECO2011-23197), the
GeneralitatdeCatalunya(2014SGR-830AGAUR),theBarcelonaGSEResearchNetwork,andthe
ERC(AdvancedGrantFP7-249588-ABEP).VothacknowledgessupportfromtheERC(Advanced
GrantFP7-230515),theUniversityofZurich,andtheSNF(100018-156197).
1
DEBT INTO GROWTH 2
1. Introduction
Over the course of a century, a country accumulates towering debts, mainly to
finance foreign wars – it is fighting abroad in two years out of three. Could such a
country transition from centuries of stagnation to sustained growth? Surprisingly,
theanswerisyes–theIndustrialRevolutioninBritainoccurredundersuchcircum-
stances. The Glorious Revolution of 1688 turned Britain into a credible borrower;
subsequently, borrowing increased massively (North and Weingast 1989). From
1692 to 1815, Britain’s debt rose from 5% to over 200% of GDP (Sussman and
Yafeh 2006; Barro 1987). The funds raised were not used to finance productivity-
enhancinginfrastructures,butinsteadtopayforoverseaswars.1 Duringthisperiod,
Britainwasatwarfor76years–62%ofthetime. Andyet,frequentwarsandhigh
debtaccumulationcoincidedwitharemarkabletransformationoftheeconomy. By
theendoftheperiod, Britain’sproductivecapacityhadgrownbyafactorofeight,
allowing it to sustain a population that was four times larger and twice as rich.
Having moved millions of people from the countryside to urban centers, it had
become the “workshop of the world”.
How could Britain industrialize while accumulating towering debts that mainly
financed foreign wars? Earlier research concluded that debt accumulation in 18th
century Britain was detrimental to industrialization since it reduced the savings
available for private investment: “Government borrowing had another ... effect.
Capital was deflected from private to public uses, and some of the developments
of the industrial revolution were once more brought to a halt” (Ashton 1948).
Williamson (1984) used a calibrated model of the British economy to show that
this crowding-out effect might have slowed output growth by as much as half of
the potential growth rate. But crowding-out should work through interest rates,
and there is little evidence that they increased.2 Barro (1987) argued that debt
accumulation had a neutral effect on industrialization since it raised total savings
instead of reducing private investment.3 Note that both Williamson and Barro
assume that the British economy possessed well-functioning credit markets.4 The
key question is then how much private savings increased in anticipation of future
taxesrequiredtoservicethedebt. Williamson’sansweris‘notmuch’whileBarro’s
view is ‘one-to-one’.
1ThisisnottodenythatsomeofthemilitarygainsthatBritainmadehadbeneficialside-effects
(suchascontrolofCaribbeanislandsproducingsugar-consumedandtaxedheavilyintheBritish
Isles). We simply argue that the overwhelming majority of expenditure resulted in no direct
economicgains.
2Research on interest rates and the yield on private assets has found few effects of sovereign
borrowing. See Mirowski (1981); Clark (2001); Quinn (2001); Sussman and Yafeh (2004). Only
Heim and Mirowski (1987) found evidence that nominal interest rates were somewhat higher
duringtheRevolutionaryWarswithFrance,buteventhentheyfoundthatrealyieldswerelower.
3Indeed, historians have noted the highly elastic supply of savings in 18th century Britain (Neal
1993).
4Bythis,wemeanthattheyassumethatno-arbitrageconditionsgenerallyheld,andthatinterest
ratesongovernmentbonds,forexample,areinformativeofthetightnessofprivatecredit.
DEBT INTO GROWTH 3
In this paper, we argue instead that Britain’s debt accumulation accelerated
industrialization. We model the Industrial Revolution as the arrival of new, high-
productivity technologies.5 Entrepreneurs invest in these new industries because
profit rates are high. Initially, entrepreneurs are relatively poor and own a small
fraction of the economy’s savings. The lion’s share of capital is in the hands of
the nobility. Earls and dukes invest in agriculture and traditional industries where
profit rates are relatively low. Ideally, entrepreneurs would borrow massively from
nobles;thiswouldleadtofastergrowthandamorerapidstructuraltransformation.
Financial frictions make his impossible: the banking sector is small and relatively
inefficient,andthestockmarketishamstrungbygovernmentrestrictions. Prejudice
also plays a role, as the nobility shied away from money-making activities. As a
result,entrepreneursareforcedtofinancetheirinvestmentsoutofreinvestedprofits;
capital formation and industrialization are relatively slow.
In such a setting, sovereign debt accelerates structural change. Sovereign bonds
are attractive to the nobility because they offer higher returns than investments in
agriculture and traditional industries. Entrepreneurs, on the other hand, are not
tempted to buy sovereign bonds because returns in the new industries are even
higher. Therefore, sovereign debt reduces investments in agriculture and tradi-
tional industries. Reduced labor demand in traditional sectors in turn depresses
wages economy-wide, raising profit rates for entrepreneurs. Since reinvested prof-
its provide most of their financing, this raises investment in new industries. In
combination, this will ensure that sovereign debt accelerates structural change and
growth. In contrast to Williamson and Barro, we emphasize the role of frictions in
private credit markets – before the 19th century, little external financing found its
way into new industries, despite huge profit opportunities. As we describe in sec-
tion II, throughout the Industrial Revolution private credit was limited, expensive,
and it provided almost no resources to the most dynamic sectors of the economy.
Our model highlights a number features whose importance has become more
apparentinrecentyears–thekeyroleofresourcereallocationandofcreditmarket
frictions in development (Hsieh and Klenow 2009, Banerjee and Moll 2010 and
Gancia and Zilibotti 2009). The effects of credit market frictions is surveyed in
BanerjeeandDuflo(2005)andhighlightedinrecentresearchby,interalia,Banerjee
and Duflo (2014) and Banerjee and Munshi (2005) . Our paper is also related to
ItskhokiandMoll(2014)whoshowthatdevelopmentpoliciesthatlowerwagesand
relaxcreditconstraintsforentrepreneurscanfostergrowth. Ouremphasisoncredit
marketfrictionsinacontextofunevengrowthindifferentsectorsisrelatedtoSong,
Storesletten,andZilibotti’s(2011)recentworkonChina. Byemphasizingresource
reallocation and credit market frictions, our model offers a unified explanation for
key macroeconomic aspects of the Industrial Revolution.
One aspect is relatively slow growth at the start of the Industrial Revolution. A
key insights from the last 30 years of research on the British economy after 1700
5ThisisinthespiritofHansenandPrescott(2002).
DEBT INTO GROWTH 4
is that growth was relatively slow before 1850, with output per capita rising at
a rate of 1% p.a. or less (Crafts and Harley 1992; Antràs and Voth 2003). As
in the seminal work by Crafts (1985), structural change is the key characteristic
of the industrialization process in our approach – the shift out of agriculture and
into industry. Our model offers one interpretation of why growth was not faster:
slowcapitalformationsloweddownstructuralchangebecauseitwaslimitedbythe
self-financing ability of entrepreneurs.6
AsecondaspectoftheIndustrialRevolutionthatourframeworkshedslightonis
thesocialchangeengenderedbytheIndustrialRevolution. Britain’snobilityin1700
held the vast majority of wealth and political power; by 1900, its relative position
had declined markedly. The nobility did not invest directly in new technologies; it
alsodidnotlendtocapitalists,eitherdirectlyorthroughthefinancialsystem. Had
the nobility been able to finance the new class of entrepreneurs in a competitive
well-functioningcreditmarket,itwouldhaveappropriatedvirtuallyalloftheprofit
opportunities arising from the new technologies – and the Industrial Revolution
would have generated little or no social change.7
A third aspect of the Industrial Revolution is limited gains in terms of living
standards accruing to the working class. Real wages did not keep up with output
growthduringthecorephaseoftheindustrializationprocess(1770-1830);thewage
share of national income fell sharply, while the share going to capital surged (Allen
2009a, Allen 2009b). Our model offers an explanation for this puzzling feature, by
showing how massive sovereign borrowing contributed to the divergence between
productivity and wages. It is precisely the reduction of labor demand (because
nobles switched from low-return investments to sovereign debt) that kept wages
low and generated the entrepreneurial profits needed to finance industrialization.
We assume that Britain’s wars generated few direct economic benefits. The
maintenance of an empire in the age of mercantilism wasfc helpful in fostering
trade (O’Brien 2000), but the claims of World System Theory about the colonial
origns of European growth are clearly misplaced (Wallerstein 2011). There is little
doubtthatlong-distancetradewastoosmalltoactasakeydriverofBritishgrowth
(Mokyr 1999).
We deliberately abstract from other aspects of the Industrial Revolution. First,
we take technological change as given. While the aggregate productivity statistics
do not show it, the 18th century saw many important inventions and innovations,
fromtheuseofsteampowertoadvancesincottonspinning,weaving,andtransport
(Mokyr1990). Nordoweseektoexplainwhytheseadvanceswerefirstconceivedor
implemented in Britain (Allen 2009b). Furthermore, we do not examine the role of
newsourcesofenergy(Wrigley1990,Stokey2001),norofforeigntrade(Crafts1985,
Harley and Crafts 2000, Temin 1997) or of improvements in transport (O´Rourke
6Crafts(2005)summarizesanalternativeviewthatseekstoexplaintheconfluenceofrapidstruc-
turalchangeandslowgrowth.
7Thisdependsontherelativebargainingpowersofsaversandinvestorsinnewtechnology. How-
ever,sincethenobilitywassmall,itislikelythatitwouldhaveexertedsubstantialmarketpower.
DEBT INTO GROWTH 5
and Williamson 2005, Bogart 2009). Finally, we do not consider the impact of
institutional improvements (North and Weingast 1989, Mokyr and Nye 2007). All
these factors undoubtedly contributed to the Industrial Revolution in ways small
or large. Here, we focus on the factors that determined how quickly technological
change made itself felt in the economy at large.
The rest of the paper is organized as follows. Section II provides the historical
background and context. It reviews the stylized facts of the Industrial Revolution
and it also describes the main features of the British financial system. Section III
presentsourmodelandderivesouranalyticalresults. Italsoprovidesaveryrough
attempt at quantification. Section IV concludes.
2. Historical Background
In this section, we first briefly summarize key macroeconomic features of the
British Industrial Revolution, as well as of the political context. In addition, we
discussthesocialanddistributionalconsequencesofthetransitiontoself-sustaining
growth, and we highlight the main features of the UK financial system.
2.1. War and the growth of debt in eighteenth-century England. The so-
called Glorious Revolution in 1688 deposed James II from the throne. Parliament
invited William of Orange to become monarch. The new constitutional settlement
includedmajorrestrictions ofthemonarch’spowers, andamuch-expandedrolefor
Parliament. Taxation required the parliamentary assent; the judicial powers of the
king were severely curtailed (North and Weingast 1989).
At the time of the Glorious Revolution, Britain had only a small national debt.
In the next 150 years, the total debt stock rose rapidly. Between 1692 and 1815,
debtrosefrom5%ofGDPtomorethan200%(Barro1987). Thecostofnumerous
wars that followed the accession of the Hanoverian kings to the throne was largely
responsible. Britainfounditselfatwarfor81years,oralmost2outofevery3. The
expenditureonthearmedforceswasconsiderable,andconstitutedbyfarthesingle
most important item of the government budget. A single ship of the line of the
Royal Navy cost more than all the capital in the most expensive iron-works built
at the time (Brewer 1990). In the period 1692-1815, spending on the Army, Navy
and on ordinance was equivalent to 72% of total revenue. Once the debt service
costs were added to this figure – due to the debts accumulated in wartime – there
was hardly any money left for non-military spending.
Figure 2.1 shows the path of overall expenditure and of total debt. Shaded
areas indicate wars. Dramatic spikes in total spending almost always coincided
with major wars. Almost the entire rise in debt during the 18th century occurred
while Britain was fighting abroad. Once peace was concluded, debt levels typically
stabilized in nominal terms, and GDP growth reduced the debt burden over time.
Peaking at over 200% after the end of the Napoleonic Wars, debt eventually fell to
100% of GDP by the middle of the 19th century.
DEBT INTO GROWTH 6
Figure 2.1. Debt and Expenditure in the UK, 1692-1860
Due to frequent wars, borrowing needs were substantial. In addition, new finan-
cial instruments facilitated the growth in public debt. Prior to the 18th century,
most borrowing by the English Crown was complex and created liquid assets. So-
calledtallies–notchedwoodensticksdenotingvariousamountsoftaxespayableto
the government – were used to borrow. In effect, tally rods acted like short-dated
IOUsissuedbythegovernment,backedbyaspecifictaxstream. Whilethesecould
be resold, trading was typically highly illiquid, with discounts of more than half of
face value. After numerous experiments, the British government granted privileges
to several companies, in exchange for financing the public debt. The most impor-
tant included the Bank of England, the New East India Company, and the South
Sea Company. All of these received royal charters in exchange for taking on some
of the government’s debts. In addition, government bonds were combined with a
nationallottery(MillionAdventure). Lifeannuitieswereissued,aswellastontines.
Short-term borrowing in case of war by the armed forces produced so-called army
and navy bills, effectively short-dated promises to pay. The biggest experiment of
all involved the South Sea Company, which offered to exchange all public debt in
1720 for shares. A similar exercise in 1719 had been attractive to both the gov-
ernment and the public, by improving the liquidity of outstanding debt. While the
SouthSeaschemeultimatelyfailed, itdemonstratedtheattractionsofliquidpaper
assets. The UK finally introduced consolidated annuities (“consols”), perpetual
bonds with a relatively low interest rate (Dickson 1967). These were first issued in
1751. Originally carrying a yield of 3.5%, they were eventually converted to 3% in
1757 (and to 2.75% in 1888). Consols were liquidly traded, and became a prime
savings vehicle for the moneyed classes in the UK.
2.2. Britain’s growth and industrial transformation. Growth during the
classic period of the British Industrial Revolution (1760-1850) was slow by modern
DEBT INTO GROWTH 7
Figure 2.2. Growth of Output in Britain, 1700-1860
standards. Initially, output growth per capita was barely faster than during the
pre-industrial period.8 After the middle of the 18th century, growth accelerated
from around 1 % p.a. to 2.5%. At the same time, population increased rapidly,
from 5.2 million to 19 million. Growth rates across sectors were highly unequal.
Figure 2.2 shows annual GDP by sector. Agriculture expanded relatively slowly
overtheperiod1700-1860, increasingtotaloutputbyafactorof2.8–aslowerrate
of increase than that of population.9 Over the same period, real GDP in services
increased 9-fold, and in industry, 14-fold (Broadberry et al. 2010) .
TFP growth remained below 1% p.a. until the 1830s, if not longer. After
1760, investment as a percentage of GDP probably increased from around 8% to
over 12%. Capital input in the economy grew slightly more rapidly than labor
input, creating a small rise in capital intensity. Land in use also increased as more
marginalareaswerebroughtundertheplough. AggregateTFPgrowthwasnothigh
by the standards of later periods, but faster than before 1760. These findings are
confirmed by studies using the productivity dual – factor returns did not increase
rapidly overall.10 Table 1 summarizes the key facts:
Wages increased relatively slowly between 1760 and 1860, by approx. 30% (Fe-
instein 1998). The remuneration of labor thus failed to grow in line with output
per capita.11 The same is true for land rents. Because rents and wages grew less
rapidly than output, the share of capital in total factor payments increased (Table
2). The period 1770-1850 therefore saw a major increase in the capital share of
8Galor(2005)givesafigureof0.1%p.a. forthepre-industrialera,whiletheworkofCraftsand
Harleysuggestsratesof0.2%p.a. intheyears1760-1800.
9Fortheeffectsofpopulationpressureoneconomicstructure,cf. CraftsandHarley(1992).
10AntràsandVoth(2003).
11MorerecentworkbyClark(2005)suggestsslightlyfastergrowth,butstillconfirmstheconclu-
sionthatoutputgrowthwasmarkedlyfasterthanwagegrowth.
DEBT INTO GROWTH 8
Table 1. Estimates of Productivity Growth in England, 1770-1860
change in % p.a.
r w q TFP
Antràs and Voth (2001)
1770-1801 -0.28 0.40 0.13 0.12
1801-1831 0.83 0.34 0.20 0.49
1831-1860 0.74 0.56 0.11 0.56
Y K L T TFP
Harley (1999)
1760-1800 1.0 1.0 0.8 0.2 0.19
1801-1831 1.9 1.7 1.4 0.4 0.50
1831-1860∗ 2.5 2.0 1.4 0.6 1.00
Sources: AntràsandVoth(2003AntrasandVoth(2003)),Harley(1999,p. 183).
Notes: r–rentalrateofcapital,w–realwage,q–rentalcostofland
Y–outputgrowth,K–capital,L–labour,T–land
AntràsandVoth(2003)useanelasticityof0.35forcapital,0.15forland,and0.5forlabour.
output (Allen 2009b), which rose from 18% to 45% in less than a century. Growth
of the capital stock was only one reason for the rise in the capital share. Crafts
(2005)estimatesthatcapitaldeepeningcontributedapproximatelyonethirdtoto-
tal per capita output growth during the Industrial Revolution. The rate of return
on capital in Britain also grew from 10% in 1770 to over 23% by 1860.
Table 2. Factor shares in Britain, 1770-1860
1770 1860 Change 1770-1860
Capital 18% 45% +27%
Labor 60% 47% -13%
Land 22% 8% -14%
The slow rate of output growth is remarkable not least because technological
change was rapid. A “wave of gadgets” (Ashton 1948) in the 1770s and 1780s
created the foundations for new products and processes. The frequency of useful
patents, as measured by their subsequent citation frequency, surged. Foreign trade
statistics suggest that technological change was broad-based.12
These changes in factor shares also spelled social change. As income on capital
increased, the relative importance of landed wealth declined. Slowly but steadily,
Britain’s nobility lost its position of paramount importance in the national wealth
and income rankings (Rubinstein 1977).
RapidstructuralchangewasthemostoutstandingfeatureofBritain’sindustrial
transformation (Crafts 1985). While total population more than doubled, the ab-
solutenumberofpeopleemployedinagricultureremainedbroadlyconstant. Thus,
the share of Britons working in agriculture fell from 57 to 25 percent over a 150
12Cf. Temin(1997). TheconclusionhasbeenquestionedbyCraftsandHarley(2000).
DEBT INTO GROWTH 9
Figure 2.3. The growth of modernizing industrial sectors, 1770-
1831
yearperiod.13 Comparedtothe‘Europeannorm’,Britainexperiencedanearlyand
rapid shift from the primary to the secondary sector (Wrigley 1985).
2.3. The rise of new industries. While aggregate growth rates were low, some
individual sectors grew rapidly. Cotton, and to a lesser extent, iron production,
transportation, and machinery expanded rapidly, starting from a low base. Pro-
ductivity in cotton production increased by 2.6% annually, shipping by 2.3%, and
worsteds and canals by 1.3 and 1.8% respectively over the period 1780-1860. This
implies that Britain became 8.8 times as productive in producing cotton goods, for
example. Output in new sectors expanded even more, as more labor and capital
were drawn in. Cotton production increased 42-fold over the period.
As the new sectors grew, their share of the industrial sector increased – and the
economyitselfbecameincreasinglyindustrialized. Figure2.3illustratesthegrowing
importance of “modernizing” sectors. From 13% of industrial output, modernizing
sectors grew to 36% by 1831. In other words, by 1830, more than one third of
industrial output already came from the sectors that benefited the most from the
new inventions of the industrial era, a share that only increased subsequently.
2.4. The UK financial system. Most historians agree that a “financial revo-
lution” preceded the Industrial Revolution (Dickson 1967). It mostly occurred in
public finance, with the introduction of liquidly traded, highly standardized gov-
ernmentdebtcontracts,andthesetting-upofataxsystemthatcouldfinancerising
debtexpenditures. Inaddition,firmsandprivateindividualscouldavailthemselves
of extensive transfer services, merchants could draw bills and discount them in a
13Crafts(1985).
DEBT INTO GROWTH 10
variety of ways, and market efficiency was generally high - meaning identical risks
were similarly priced (Neal 1994).
At the same time, the UK financial system was not very effective in mobilizing
resources to the new industries. As Larry Neal observed “...the industrial revolu-
tion[took]place... withoutanyinteractionwiththeprecedingfinancialrevolution”
(Neal 1994). In other words, “the reservoirs of savings were full enough, but con-
duits to connect them with the wheels of industry were few and meagre ... surpris-
inglylittleof[Britain’s]wealthfounditswayintothenewindustrialenterprises...”
(Postan 1935) . The new, dynamic sectors of the economy – cotton manufactur-
ing, iron production, coal mining, ceramics – largely financed themselves through
retained profits and informal credit. Peer-to-peer lending dominated where credit
was used at all – entrepreneursturned to friends, family, and local owners of liquid
funds to raise capital (McCraw 1997; Mokyr 1999).14
While private credit markets worked well in many ways, they were poor at mo-
bilizing funds for industrialization for several reasons. The Bank of England was
largely a conduit for government debt. Goldsmith banks were small and few in
number,andonlycateredtoamoneyedandlandedelite. Merchantbanksfinanced
foreign trade. Almost no financial institutions, except a few country banks, at-
tempted to provide funds for entrepreneurs (Pressnell 1956).15In part, this was
because banks were hamstrung by government regulations. The size of partner-
ships in England was limited to six, severely curtailing the size of banks. Usury
laws limited (private) interest rates. This rule made lending on anything but the
best collateral, and to the safest borrowers, unprofitable. Temin and Voth (2008)
document how the usury law led to massive distortions in lending. In addition,
loans could legally not be made for periods greater than six months. This made
it hard for borrowers to use funds for illiquid investments. There was no central
bank, charged with providing liquidity. Existing banks struggled under the threat
of illiquidity, and many floundered because of it.
Nor could new firms easily raise equity on the stock market. As a result of the
South Sea bubble, the government introduced tight restrictions on the founding of
newjointstockcompaniesintheformoftheso-called“BubbleAct”. Thelegislation
required all new joint stock companies to have a royal charter. Effectively, until its
repeal in the 19th century, the Bubble Act closed the door on all forms of capital
raising via the issuance of new equity (Harris 1994). In combination, private credit
intermediation in Britain before the 1820s worked poorly at best. On the whole, it
14IncontrasttoFrance,Britainhadnosystemofpublicnotarieswhofacilitatedsuchtransactions
(Hoffman,Postel-Vinay,andRosenthal2000). Notethatinformallendingisstillcommontoday,
bothinthedevelopingworldandincountrieswithhighlydevelopedcapitalmarkets(Azametal.
2001,Townsend2005).
15TeminandVoth(2013). Brunt(2006) arguesthatcountrybanksactedlikeventurecapitalfirms
during the British Industrial Revolution. The cases he shows are suggestive, but it is doubtful
thatthisrepresentsanimportantpartofbusinessfinancingatthetime. Theanalogywithventure
capitalfirmsisalsostrained,sincetheupsidetothebankwasseverelylimited–atnomorethan
5%.